| KEYWORD |
Area Engineering
Generative AI for Generation of OpenScenario and OpenDrive from series of traffic images
Thesis in external company
keywords AI, AUTOMOTIVE, TESTING
Reference persons RICCARDO COPPOLA
Research Groups DAUIN - GR-16 - SOFTWARE ENGINEERING GROUP - SOFTENG
Thesis type COMPANY STAGE
Description Context:
In the realm of Advanced Driver Assistance Systems (ADAS) and Autonomous Vehicles (AV), drive and traffic simulators play a crucial role in accelerating development, training, and validation processes. This dynamic field is witnessing a growing demand for standardized simulation models, aiming for the seamless exchange of data and models.
Description:
The candidate will join a dynamic Research and Innovation team dedicated to pushing the boundaries of data-driven development. This role focuses on developing a cutting-edge Software as a Service (SaaS) platform, seamlessly translating traffic images into OpenSCENARIO and OpenDRIVE files using Large Language Models.
Objectives:
Explore state-of-the-art techniques, large language models, and structured text generation
Prompt engineering for optimized generation of standardized files from traffic images
Fine-tune ChatGPT for a customized model
Design and execute output validation using CARLA simulator / esmini testing environment
Develop end-to-end service, including customer journey definition, front-end development, ChatGPT integration, and deployment on a physical server
Required skills Pre-requisites:
Experience with OpenAI ChatGPT (or similar) API integration for generating structured textual descriptions of images
Demonstrated interest in Autonomous Driving
Full-stack web development skills
Deadline 26/02/2025
PROPONI LA TUA CANDIDATURA